Project 2-Summary There is heterogeneity in many aspects of peanut allergy (PA), including the dose of peanut that triggers symptoms (the threshold), the severity of symptoms that are induced, and the type of symptoms (target organ(s)) that are triggered. A significant minority (up to 20%) will outgrow their PA, while the majority retains clinical reactivity. We propose that this natural heterogeneity in the clinical phenotype of PA is a valuable opportunity to understand immune mechanisms contributing to disease. We hypothesize that distinct immune responses at the level of IgE, IgG4, CD4+ T cells, and basophils underlie important clinical heterogeneity, and furthermore we hypothesize that these distinct immune profiles form a basis for prognosis of PA. Through the Clinical Core, we will have access to 200-250 subjects undergoing oral food challenge for PA who have a low or high threshold of reactivity to peanut, or who are not allergic. We will also have access to 98 high-threshold subjects randomized to receive sub-threshold peanut in the diet (or continue avoidance), and who will have a diverse set of clinical outcomes over the 2.5 years that they are studied. We will perform high dimensional immune profiling on these subjects to identify pathways linking immune and clinical responses to peanut. We have found that bioinformatics analysis of IgE binding to epitopes in milk allergens can predict not only clinical phenotype of milk allergy, but accurately predict treatment success after milk oral immunotherapy. We will apply this approach to test if epitope binding can predict clinical reactivity to peanut, and predict response to dietary intervention. We have also identified that the magnitude and differentiation of the peanut-specific Th2 response is highly heterogeneous in PA subjects. We will determine the clinical implications of Th2 heterogeneity within this diverse cohort, using a combination of CD154-based detection and phenotypic analysis of peanut-responsive T cells by flow cytometry, and multiplex cytokine assays. We have developed CyTOF-based approaches to study the response of cells in whole blood to peanut, and have published as well as preliminary data demonstrating that we can identify signaling and activation in basophils and other cell types not only in vitro but also in vivo. We hypothesize that broad profiling of the activation of cells in whole blood by CyTOF will identify novel communities of responding cells that discriminate between key clinical phenotypes of PA. The successful completion of our project will build a clinical-immune network of PA that overlays a detailed clinical phenotype over a high dimensional network of peanut-specific antibodies, T cells, and effector cell responses. We anticipate that this clinical-immune network will not only reveal mechanisms, but will be an essential tool for guiding personalized management and treatment strategies for PA.